TYP: annotations in pandas.plotting #35935
Conversation
| from pandas.core.dtypes.common import is_list_like | ||
| from pandas.core.dtypes.generic import ABCDataFrame, ABCIndexClass, ABCSeries | ||
|
|
||
| from pandas.plotting._matplotlib import compat | ||
|
|
||
| if TYPE_CHECKING: | ||
| from matplotlib.table import Table |
There was a problem hiding this comment.
Revealed type is 'Any'
not sure what's best here. I imagine that once we get np.ndarray imports sorted, there'll be a lot of mypy errors to fix. I wonder whether we want to continue adding types that currently resolve to Any or leave untyped.
There was a problem hiding this comment.
ill follow your lead on this.
there are a lot of places in pd.plotting where we have a ax arg that would be nice to disambiguate pd.Index vs matplotlib.axes.Axes vs matplotlib.axis.Axis (i think most are Axes)
There was a problem hiding this comment.
once we get np.ndarray imports sorted
is there cause for optimism this may be soon-ish?
There was a problem hiding this comment.
I'm not sure that we have got to any technical blocks yet since not much effort in this direction. Personally, I wanted to get the check_untyped_defs sorted first. (was a phase 0 in a plan) that imo comes before all other work. but in open source can't control the effort and therefore some have a preference for typing the public api. My interest in typing has always been for the internal consistency and quality of the pandas codebase itself.
But with hindsight, getting typing PRs through was so painfull that any typing additions is a bonus.
|
Thanks @jbrockmendel |
No description provided.